An explorative multiverse study for extracting differences in P3 latencies between young and old adults

Authors:
  • KSKathrin Sadus
    Kathrin Sadus
  • ASAnna-Lena Schubert
    Anna-Lena Schubert
  • CLChristoph Löffler
    Christoph Löffler
  • DHDirk Hagemann
    Dirk Hagemann

Abstract
It is well established that P3 latencies increase with age. Investigating these age-related differences requires numerous methodological decisions, resulting in pipelines of great variation. The aim of the present work is to investigate the effects of different analytical pipelines on the age-related differences in P3 latencies in real data. Therefore, we conducted an explorative multiverse study and varied the low-pass filter (4 Hz, 8 Hz, 16 Hz, 32 Hz, and no filter), the latency type (area vs. peak), the level of ERP analysis (single participant vs. jackknifing) and the extraction method (manual vs. automated). 30 young (18 – 21 years) and 30 old (50 – 60 years) participants completed three tasks (Nback task, Switching task, Flanker task) while an EEG was recorded.The results show that different analysis strategies can have a tremendous impact on the detection and magnitude of the age effect, with effect sizes ranging from 0% to 88% explained variance. Likewise, regarding the psychometric properties of P3 latencies, we found that the reliabilities fluctuated between rtt = .20 and 1.00, while the homogeneities ranged from rh = -.12 to .90. Based on predefined criteria, we found that the most effective pipelines relied on a manual extraction based on a single participant's data. For peak latencies, manual extraction performed well for all filters except for 4 Hz, while for area latencies, filters above 8 Hz produced desirable results. Furthermore, our findings add to the evidence that jackknifing combined with peak latencies can lead to inconclusive results.